We peer back hundreds of millions of years into protein’s past with AlphaFold to learn more about the origins of life itself
Pedro Beltrao is a geneticist at ETH Zurich in Switzerland. He shares his AlphaFold story.
As a scientist, I am interested in our differences.
More specifically, I’m interested in how these differences arise. While many people work on how changes in DNA cause changes in our traits—for example, from predisposition to certain diseases, or simply why some people are taller than others—our research examines why this happens.
Ultimately, we would like to have a model that tells us exactly how a person will change or what characteristics they will have if they have a mutation at a specific location in their DNA.
There is still a long way to go to achieve this.
The first layer involves figuring out which mutations in DNA don’t cause any change. To do that, you have to ask yourself: does this affect proteins or not? Then, because proteins work together to perform functions, we need to know how that works and how those functions present themselves. That can mean different things depending on whether we’re considering a brain cell, a kidney cell, or a skin cell. Of course, each organ is different, too. There are many progressions and variables to understand, from a single mutation to a protein, to a group of proteins, to just cellular tissue, and then figuring out how the whole organism behaves.
Before AlphaFold, we had some protein structures for both individual proteins and complexes—probably about 5% of the pairs that interact had a known structure, for example. Now that’s changing rapidly. What’s more, we now have an electrifying opportunity to study protein evolution in the origin of life.
I find this part of our research particularly electrifying. When we want to go back in time to look at evolution, we usually do it by comparing sequences between proteins in different species. That way we can try to guess what that sequence looked like in the evolutionary past.
Without protein structures, we can only go back in time so far: there comes a point where we lose confidence in how things were hundreds of millions of years ago. Using AlphaFold and comparing the 3D shape of proteins, it preserves the signal for a longer time, because the 3D structure of proteins is preserved longer than the sequence that encodes that shape.
This allows us to now trace the evolution of proteins back over longer periods of evolutionary timescales, and to infer with greater confidence what the earliest ancestral cell looked like by analyzing what proteins looked like hundreds of millions of years in the past.
Often in science we have these incremental changes where modern technologies, methods or systems build up over time or evolve slowly. And every so often we have periods of transformation. There’s no doubt that AlphaFold has triggered a period of transformation. It’s incredibly electrifying. Now we have the opportunity to learn so much more about human biology and the origins of life itself.